Detection and identification of single dye labeled mononucleotide molecules released from an optical fiber in a microcapillary: First steps towards a new single molecule DNA sequencing technique

Abstract

We present efficient detection of single dye labeled mononucleotide molecules released from an optical fiber in a cone shaped microcapillary with an inner diameter of 0.5µm at the small end. A 3µm diameter streptavidin coated optical fiber was doped with dye labeled mononucleotide molecules and positioned about 400µm in front of the detection area. The flow of the negatively charged analyte molecules towards the detection area was established by electrokinetic forces. Adsorption of the dye labeled mononucleotide molecules on the capillary walls as well as the electroosmotic (EOF) flow could be drastically reduced using a 3% polyvinylpyrrolidone (PVP) matrix. We studied bursts of fluorescence photons from single fluorescently labeled mononucleotide molecules Cy5-dCTP and MR121-dUTP. The conjugates were excited by a short-pulse diode laser emitting at a wavelength of 638 nm with a repetition rate of 57 MHz. The setup consists essentially of a confocal microscope for time-correlated single-photon counting (TCSPC). Detection was performed by a single-photon avalanche-photodiode in combination with a PC plug-in card for TCSPC. The time-resolved fluorescence signals of individual molecules were analyzed and identified by a maximum likelihood estimator (MLE). We obtain an overall classification probability of about 96% for the time-resolved identification of Cy5-dCTP and MR121-dUTP via their characteristic fluorescence decay times of 1.42±0.21 ns (Cy5-dCTP) and 2.41±0.32 ns (MR121-dUTP). In addition we report on detection and identification of single Cy5-dCMP and MR121-dUMP molecules degraded enzymatically with the 3′→5′ proofreading activity of T4-DNA-polymerase from statistically labeled DNA strands bound to the optical fiber. The technique presented permits counting and identification of fluorescently labeled mononucleotide molecules cleaved from DNA strands with counting rates>200 nucleotides s^-1 and represents a suitable alternative for single molecule sequencing.